readable AI benchmarks (simplified)
Sarvam
Sarvam 105B (High)
Sarvam 105B (High) has a Quality Score of 16.9, ranking 464th among 571 scored models. Per 1M tokens, pricing is $0.042 input, $0.170 output, and $0.026 cached input. It accepts text input, outputs text, and has a 128k token context window. Its Reliability Score is 20.3, ranking 474th among 573 scored models, below average (total average: 36.5). Its Value Score is 16.0, low compared with the total Value average of 31.4.
Quality Score16.9coverage 100.0%: AA Intelligence and AA-Omniscience outcomes available
Value Score16.0
Factual reliability20.3
Cache Discount38%
Model specification
- Reasoning
- reasoning
- Input modalities
- text
- Output modalities
- text
- Context window
- 128k tokens
- Weights
- open weights
Token prices USD per 1M tokens
- Input
- $0.042
- Output
- $0.170
- Cached Input
- $0.026
- Cache write
- not available
- Cost per task
- not available
Capability
- Intelligence
- 8.8
- Coding
- not available
- Agentic
- not available
- Omniscience
- -59.4
- Correct
- 17.6%
- Blended price
- $0.044
Answer outcomes
Fully graded outcomes (Correct + Incorrect): 94.6%
Correct 17.6%Incorrect 77.0%Partial / not attempted 5.4%
Artificial Analysis benchmarks
- Humanity’s Last Exam
- 11%
- GPQA Diamond
- 74%
- CritPt
- 0%
- AA-Omniscience
- 20%
- AA-LCR
- 0%
Similar models
- Sarvam M (Reasoning, Based on Mistral Small 3.1)SarvamQuality Score 13.7Value Score not available
- Sarvam 30B (High)SarvamQuality Score 12.2Value Score 11.7
- Sarvam MSarvamQuality Score not availableValue Score not available
- Qwen3 4B 2507 (Reasoning)AlibabaQuality Score 16.7Value Score not available
- Llama 3.3 Nemotron Super 49B v1 (Reasoning)NVIDIAQuality Score 17.3Value Score not available
- Motif-2-12.7B-ReasoningMotif TechnologiesQuality Score 16.6Value Score not available